Recently, communication systems performing communication via communication networks such as the Internet and exclusive lines have become widespread along with requests for reductions in business trip expenses and business trip times. Such a communication system establishes a session for transmitting and receiving content data such as image data and voice data between communication terminals, thereby achieving communication between the communication terminals.
It is known that, when communication between communication terminals is started, a communication system of a teleconference controls a session by a teleconference management system (refer to Patent Literature 1). In this case, a plurality of teleconference terminals are connected to one teleconference management system, and based on a request from any of the teleconference terminals, the teleconference management system performs processing related to the start of communication. Owing to this, when the number of the teleconference terminals connected to the teleconference management system increases, a large load associated with processing is imposed on the teleconference management system.
A method is known that, in order to reduce a load imposed on a server, provides a plurality of servers to perform processing, thereby distributing the load imposed on the server (refer to Patent Literature 2).
However, when a plurality of controllers are provided in order to perform processing related to the start of communication between communication terminals, a problem arises in that when plural requests to start communication for the same communication terminal are respectively accepted by different controllers at the same time, the different controllers perform processing according to the requests in a duplicate manner.